SECTION 5.3 DIURNAL AND RESTING LOSS EMISSIONS

5.3.1 Introduction

While most hydrocarbons are emitted from the tailpipe of vehicles as the product ofincomplete combustion, a significant amount of hydrocarbons emanate throughevaporation, namely, hot soak, diurnal, resting and running loss emissions. Diurnaland resting loss emissions occur as a result of the vehicle’s fuel heating andvolatilizing as the ambient temperature rises or declines during the day. To reduceevaporative hydrocarbon emissions, vehicles are equipped with a canistercontaining activated charcoal to adsorb the vapors. When the vehicle is operated,the canister is purged with ambient air and the stored hydrocarbon is burned in theengine.

The diurnal emission factors in MVEI7g1.0c were developed in 1992 using datafrom Air Resources Board’s past In-Use Vehicle Surveillance programs as well asInspection and Maintenance projects. The diurnal basic emission rates wereestimated by measuring hydrocarbon vapor from a vehicle during a compressedone-hour Sealed Housing Evaporative Determination (SHED) test. During this test,a heating blanket is placed in contact with the vehicle’s fuel tank to heat the fuelfrom 60 to 84 F over a one-hour period, representing a temperature excursion thatstationary vehicles may experience daily. The resting loss basic emission rates werebased on limited 3-day and 8-day diurnal test data (SAE 901110).

Although the above methods attempt to quantify diurnal and resting loss emissions,the compressed nature of the test may not faithfully reproduce the real-timeevaporative emission process and may, therefore, underestimate emissions. In otherwords, the one-hour condensed diurnal test may not generate the total diurnal andresting loss emissions from a 24-hour diurnal. Thus, there is a critical need tomodel diurnal and resting loss emissions using real-time data to improve theemissions inventory. In this study, diurnal and resting loss results were analyzedfrom four databases: ATL(A096-214 and A132-183), CRC and EPA. All datawere collected under real-time conditions ranging from 24 hours to 72 hours withdifferent combinations of fuel and temperature cycle. Additionally, the temperatureprofiles are for the ambient SHED temperature rather than the tank temperature(with a heating blanket). 5.3.2 Objectives

This analysis has the following objectives:

1. To develop the diurnal and resting loss basic emission rates as a function ofambient temperature and fuel RVP.

2. To develop the fuel and temperature correction factors.3. To develop the multi-day correction factors for diurnal and resting loss

emissions

4. To define evaporative emission regimes (normal, moderate, and liquid leaker)and develop regime growth rates for CARB and FI vehicles.

5.3.3 Methodology

Figure 5.3-1 outlines the methodology used for the analysis of diurnal and restingloss emissions. The four databases were combined into one. The resulting data areshown in Table 5.3-1a - 291 vehicles covering passenger cars and trucks withmodel years ranging from 1971 to 1995.

Figure 5.3-1. Flowchart of the methodology used in data analysis of diurnaand resting loss emissions.

Cutpoint for Normal Emitters: 15.3 g/day

High Emiters (Liquid Leakers) were identified frominspection report

ATL (A096-214)ATL (A132-183)

CRC

Corrected to 9 psi RVP

Model Year GroupingPre79

1979 - 941995 and beyond (Enhanced Evap.)

Liquid Leakers

Resting Loss Emissions Modeling

Develop emission regime growth rate

EPA

Normal Emitters

Assess the impact of I/M on emission regime growth rateAssess the impact of OBD II on emission regime growth rate

Fleet averaged diurnal and resting loss emission factors

Diurnal Emissions Modeling

Moderate Emitters

Develop fuel and temperature correction equation

Table 5.3-1b summarizes the fuel RVP and temperature cycles used in eachdatabase. With the exception of the CRC test program, most vehicles were testedover different temperature cycles using different fuels.

All vehicles were segregated into three emission regimes, namely; normal,moderate and liquid leakers. A liquid leaker was defined as a vehicle classified byits emission rate as a liquid leaker when undergoing visual inspection. Note thatmost of the liquid leakers were identified from the CRC study, which included adetailed inspection report on the condition of each vehicle.

Table 5.3-1a. Distribution of vehicles' model years from four databases.

ATL096-214 (n=10) ATL132-183 (n=11) Combined Databases (n=291)Vehicle Type Car Vehicle Type Car Vehicle Typ Car TruckModel Year CARB FI Model Year CARB FI Model Year CARB FI CARB FI

73 1 0 73 1 0 71 2 0 2 076 1 0 76 1 0 72 6 0 1 078 1 0 78 1 0 73 6 0 3 079 1 0 83 1 0 74 6 0 1 083 1 0 84 1 0 75 3 0 2 085 1 0 85 1 0 76 14 0 2 086 0 1 86 0 1 77 8 0 4 087 0 1 87 0 1 78 2 0 0 090 0 2 89 0 1 79 2 0 0 0

Subtotal 6 4 90 0 2 80 2 0 3 0Subtotal 6 5 81 3 0 7 0

82 0 0 7 083 4 0 7 0

CRC (n=151) EPA (n=119) 84 3 1 13 1Vehicle Typ Car Truck Vehicle TypCar Truck 85 6 8 15 2Model Year CARB FI CARB FI Model Year CARB FI CARB FI 86 2 13 3 4

71 2 0 1 0 71 0 0 1 0 87 3 15 4 572 6 0 1 0 72 0 0 0 0 88 2 5 0 1273 4 0 3 0 73 0 0 0 0 89 1 7 0 1474 4 0 1 0 74 2 0 0 0 90 0 12 0 875 3 0 1 0 75 0 0 1 0 91 0 11 0 776 11 0 2 0 76 1 0 0 0 92 0 4 0 077 8 0 4 0 77 0 0 0 0 93 0 6 0 378 0 0 0 0 78 0 0 0 0 94 0 3 0 079 0 0 0 0 79 1 0 0 0 95 0 0 0 180 0 0 2 0 80 2 0 1 0 Total 75 85 74 57 81 0 0 5 0 81 3 0 2 082 0 0 6 0 82 0 0 1 083 0 0 7 0 83 2 0 0 084 0 0 12 1 84 2 1 1 085 0 0 15 2 85 4 8 0 086 0 0 3 4 86 2 11 0 087 0 0 4 3 87 3 13 0 288 0 0 0 9 88 2 5 0 389 0 0 0 12 89 1 6 0 290 0 0 0 8 90 0 8 0 091 0 0 0 7 91 0 11 0 0

Subtotal 38 0 67 46 92 0 4 0 093 0 6 0 394 0 3 0 095 0 0 0 1

Subtotal 25 76 7 11

Prior to establishing a cutpoint for normal and moderate emitters, a correlationequation was developed relating the one-hour condensed diurnal and the 24-hourreal-time diurnal test (see Appendix 5.3-1). To be consistent with the cutpoint usedin the hot soak analysis, the same 2 g/test (based on the one-hour condensed diurnaltest) was selected to distinguish normal and moderate emitters. According to thecorrelation equation mentioned earlier, 2 g/test (one-hour) corresponds to 15.3g/day. Therefore, vehicles were divided into normal and moderate emitters basedon this 15.3 g/day cutpoint.

The domain of the database is described below:i Fuel RVP (6.3 psi, 7.0 psi, 7.5 psi, 9.0 psi)i Temperature cycles (60 – 84 F, 65 – 105 F, 72 – 96 F, and 82 – 106 F)i Vehicle Types (Passenger Car and Light Duty Truck)i Model Year Range (1971 to 1995)i Fuel Delivery System (CARB, TBI, and PFI)i Emission Status (Normal, Moderate and Liquid Leakers)

Table 5.3-1b. Distribution of diurnal tests from four databases.ATL096-214 (n=30)Vehicle Type CarRVP 9 psiTemp Range CARB FI

60 - 84 F 12 865 - 105 F 8 2

ATL132-183 (n=40)Vehicle Type CarRVP 6.6 psi 7.5 psi (ethanol) 7.6 psi 8.7 psiTemp Range CARB FI CARB FI CARB FI CARB FI

65 - 105 F 6 5 2 4 7 5 6 5

CRC (n=151)Vehicle Type Car TruckRVP 6.8 psi 6.8 psiTemp Range CARB FI CARB FI

72 - 96 F 38 0 67 46

EPA (n=560)Vehicle Type CarRVP 6.3 psi 6.7 psi 6.9 psi 9.0 psiTemp Range CARB FI CARB FI CARB FI CARB FI

60 - 84 F 0 6 0 0 12 43 19 6572 - 96 F 0 6 12 29 12 44 19 6482 - 106 F 0 6 7 18 12 45 13 49

(cont'd)Vehicle Type TruckRVP 6.3 psi 6.9 psi 9.0 psiTemp Range CARB FI CARB FI CARB FI

60 - 84 F 0 0 0 2 7 1172 - 96 F 7 9 0 2 7 1182 - 106 F 8 9 0 2 0 4

Since not all data were collected under the same temperature cycle and fuel RVP,the data were adjusted to a common RVP prior to analysis. To do so, the hourlyaverage emissions were computed combining all temperature cycles at each fuelRVP (6.3 RVP, 7.0 RVP, 7.5 RVP and 9.0 RVP). The hourly average emissionswere then normalized (treating the data at 9 RVP as unity). As a result, all hourlyaverage emissions of various fuel RVPs were standardized to 9 RVP (liquid leakersare not assumed to be effected by RVP).

Model year grouping were based on the same model year split used in the hot soakanalysis. Duncan’s test was performed on the data and it was found that nostatistical difference exists between: (1) car and truck and (2) TBI and PFI. As aresult, car and truck were combined into a single category, and TBI and PFI datawere combined into a single category, called Fuel Injection (FI).

Finally, the data were stratified into several categories according to emission status(normal, moderate and liquid leakers), fuel delivery system (FI and CARB), andmodel year group (Pre-79, 1979-94, and 1995 and beyond). Hourly averageemissions were then computed for each stratum. (See Appendix 5.3-2.)

5.3.4 Diurnal and Resting Loss Basic Emission Rates

Diurnal emissions are defined as the evaporative emissions occuring when theambient temperature rises. Figure 5.3-2a shows an example of how temperaturecycle relates to hourly average emissions for a particular category (normal emitter,FI, and 1979-1994). As seen in this figure, hourly average emissions rise as theambient temperature increases. Only the data corresponding to a rising ambienttemperature were used for the diurnal emission analysis. In other words, the hourlyaverage emissions from the four temperature cycles were used to develop a modelusing the ambient temperature as the independent variable. Several regressionmodels were tried and it was concluded that a third order polynomial equation fitthe data adequately.

Resting loss emissions are defined as the evaporative emissions occuring when theambient temperature (T) declines or remains constant. Only the data correspondingto the declining ambient temperature were used in the modeling of resting loss.Similar to diurnal emission, a third order polynomial equation was used. Thegeneral form of this polynomial equation is listed below.

Diurnal or Resting Loss (g/hr) = α(T) + β(T)2 + χ(T)3 + Intercept (5.3-1)

For normal and liquid leakers, it was assumed that diurnal or resting loss emissionsrates increase from 55 F to 110 F and that diurnal and resting loss emissions fall tozero at 55 F or below. While a third order equation was used to predict diurnal andresting loss emission rates reasonably for the temperature range between 65 F to110 F, a linear model was used to depict the emission rates between 55 F and 65 Ffor diurnal and resting loss. For liquid leaker, it was assumed that the diurnal andresting loss emissions rates increase from 40 F to 110 F. A third order equation wasused to predict the diurnal and resting loss emission rates between 70 to 110 F and alinear model was used to depict the emission rates from 70 F to 40 F. Table 5.3-2aand 5.3-2b list the coefficients of the regression models for diurnal and resting lossemissions for the temperature ranges described above. Any diurnal or resting lossat a different fuel RVP should be adjusted by fuel and temperature correctionfactors

Figure 5.3-2a. Average hourly emission for the category: moderate emitters/FI/model year 1979-94. T - Temperature E - Emissions

0

20

40

60

80

100

120

140

160

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Hour

Tem

p (F

)

0

2

4

6

8

10

12

14

16

g/hr

T(65-105 F)T(60-84 F)T(72-96 F)T(82-106 F)E(60-84 F)E(65 - 105 F)E(72 -96 F)E(82 -106 F)

Diurnal Emissions Resting Loss Emissions

Table 5.3-2a Coefficients of the regression model for diurnal and evaporative emissions (fuel RVP = 9 psi)*.

Diurnal Evaporative Emission (g/hr)Status System MY Group Intercept Temp Temp2 Temp3 Conditions Tech GroupNormal CARB Pre77 0.3702 -0.0220910 0.0003170 Ranges from 65 F to 110 F 1-3, 21, 22

CARB 77+ 1.3300 -0.0495310 0.0004930 Ranges from 65 F to 110 F 4,23,24

FI 79-94 -3.6979 0.1310920 -0.0015340 0.0000066 Ranges from 65 F to 110 F 5-12, 25-32FI Enhanced -0.4230 0.0149969 -0.0001755 0.0000007 Ranges from 65 F to 110 F 13, 33FI Zero Evap -0.1058 0.0037492 -0.0000439 0.0000002 Ranges from 65 F to 110 F 14, 34

Moderate CARB Pre77 -65.1714 2.5978250 -0.0346650 0.0001590 Ranges from 65 F to 110 F 1-3, 21, 22CARB 77+ -40.4512 1.5929020 -0.0208880 0.0000952 Ranges from 65 F to 110 F 4,23,24

FI 79-94 11.4632 -0.3342420 0.0026300 Ranges from 65 F to 110 F 5-12, 25-32FI Enhanced 1.3114 -0.0382 0.0003009 Ranges from 65 F to 110 F 13, 33FI Zero Evap 0.3278 -0.0096 0.0000752 Ranges from 65 F to 110 F 14, 34

High All All 25.0075 -0.6909750 0.0054520 Ranges from 70 F to 110 F All

Resting Evaporative Emission (g/hr)Status System MY Group Intercept Temp Temp2 Temp3 Conditions Tech GroupNormal CARB Pre77 2.6605 -0.0864800 0.0007370 Ranges from 65 F to 110 F 1-3, 21, 22

CARB 77+ 2.8687 -0.0870240 0.0006820 Ranges from 65 F to 110 F 4,23,24

FI 79+ 1.5166 -0.0459490 0.0003580 Ranges from 65 F to 110 F 5-12, 25-32FI Enhanced 0.1735 -0.0052566 0.0000410 Ranges from 65 F to 110 F 13, 33FI Zero Evap 0.0434 -0.0013141 0.0000102 Ranges from 65 F to 110 F 14, 34

Moderate CARB Pre77 8.0181 -0.2481270 0.0020160 Ranges from 65 F to 110 F 1-3, 21, 22CARB 77+ -37.7714 1.5544770 -0.0211460 0.0000960 Ranges from 65 F to 110 F 4,23,24

FI 79+ -9.9635 0.4569720 -0.0070080 0.0000361 Ranges from 65 F to 110 F 5-12, 25-32FI Enhanced -1.1398 0.0522776 -0.0008017 0.0000041 Ranges from 65 F to 110 F 13, 33FI Zero Evap -0.2850 0.0130694 -0.0002004 0.0000010 Ranges from 65 F to 110 F 14, 34

High All All 16.9159 -0.4379580 0.0033520 Ranges from 70 F to 110 F All

*The model is defined as:Emission (g/hr) =a*T + b*T2 +c*T3 + intercept

Table 5.3-2b. Coefficients of the linear model for diurnal and evaporative emissions (fuel RVP = 9 psi)*.

Diurnal Evaporative Emission (g/hr)Status System MY Group Temp Conditions Tech GroupNormal CARB Pre77 0.02736 Ranges from 65 F to 55 F 1-3, 21, 22

CARB 77+ 0.01934 Ranges from 65 F to 110 F 4,23,24

FI 79+ 0.01413 Ranges from 65 F to 110 F 5-12, 25-32FI Enhanced 0.00162 Ranges from 65 F to 110 F 13, 33FI Zero Evap 0.00040 Ranges from 65 F to 110 F 14, 34

Moderate CARB Pre77 0.08930 Ranges from 65 F to 55 F 1-3, 21, 22CARB 77+ 0.09818 Ranges from 65 F to 55 F 4,23,24

FI 79+ 0.08493 Ranges from 65 F to 55 F 5-12, 25-32FI Enhanced 0.00972 Ranges from 65 F to 55 F 13, 33FI Zero Evap 0.00243 Ranges from 65 F to 55 F 14, 34

High All All 0.11180 Ranges from 70 F to 40 F All

Resting Evaporative Emission (g/hr)Status System MY Group Temp Conditions Tech GroupNormal CARB Pre77 0.01531 Ranges from 65 F to 55 F 1-3, 21, 22

CARB 77+ 0.00936 Ranges from 65 F to 55 F 4,23,24

FI 79-94 0.00424 Ranges from 65 F to 55 F 5-12, 25-32FI Enhanced 0.00049 Ranges from 65 F to 55 F 13, 33FI Zero Evap 0.00012 Ranges from 65 F to 55 F 14, 34

Moderate CARB Pre77 0.04075 Ranges from 65 F to 55 F 1-3, 21, 22CARB 77+ 0.03046 Ranges from 65 F to 55 F 4,23,24

FI 79-94 0.00341 Ranges from 65 F to 55 F 5-12, 25-32FI Enhanced 0.00039 Ranges from 65 F to 55 F 13, 33FI Zero Evap 0.00010 Ranges from 65 F to 55 F 14, 34

High** All All 0.08945 Ranges from 70 F to 40 F All*The linear model for normal and moderate emitters is defined as:Emission (g/hr) = a* (T-55)

** The linear model for high emitters is defined as:Emission (g/hr) = a* (T-40)

Figure 5.3-2b presents the estimated diurnal emissions for both moderate andnormal emitters. As expected, the emission rates are higher for older vehicles.Figure 5.3-2c shows the estimated resting loss for both moderate and normalemitters. A similar trend showing higher emission rates for older vehicles was alsoobserved. When comparing Figure 5.3-2b with 5.3-2c, it was found that themagnitude of emissions from the diurnal is higher than resting loss. Figure 5.3-2dshows the diurnal and resting losses for the liquid leakers. As expected, the liquidleaker has the highest hourly diurnal and resting loss emissions.

Figure x-2b. Relationship between estimated diurnal emissions and ambient temperaturefor normal and moderate emitters.

Estimated Diurnal Evaporative Emissions (Moderate Emitters)

0

2

4

6

8

10

12

14

16

18

40 50 60 70 80 90 100 110 120Temp (F)

g/hr

M/CARB/Pre79M/CARB/79-94M/FI/79-94

Estimated Diurnal Evaporative Emissions (Normal Emitters)

0.0

0.5

1.0

1.5

2.0

2.5

40 50 60 70 80 90 100 110 120

Temp (F)

g/hr

N/CARB/Pre79N/CARB/79-94N/FI/79-94N/FI/Enhanced

Figure x-2c. Relationship between estimated resting loss emissions and ambient temperaturefor moderate and normal emitters.

Estimated Resting Evaporative Emissions (Moderate Emitters)

0

1

2

3

4

5

6

7

8

40 50 60 70 80 90 100 110 120Temp (F)

g/hr

M/CARB/Pre79M/CARB/79-94M/FI/79-94

Estimated Resting Evaporative Emissions (Normal Emitters)

0

1

1

2

2

3

3

40 50 60 70 80 90 100 110 120

Temp (F)

g/hr

N/CARB/Pre79N/CARB/79-94N/FI/79-94N/FI/Enhanced

5.3.5 Basic Emission Rate for model year 1995 and beyond (EnhancedEvaporative Emission Standards)

The enhanced evaporative emission standards for passenger cars and trucks is 2gram/test (based upon a three-day diurnal and a one-hour hot soak test), with thetemperature ranging from 65 F to 105 F. Four enhanced evap vehicles wereanalyzed from CRC Project E-41, Real World Evaporative Testing of Late ModelIn-Use Vehicles, McClement, Hall and Strunck, October 1999. The mean emissionsof these four vehicles are assumed to be representative of Normal emittingenhanced evap vehicles. To derive temperature based BERs for diurnal and restinglosses for model years 1995+, diurnal and resting loss emission rates from thecategory (Normal/FI/ 79-94) were used as a basis for estimation. The BERs for theNormal/79-94 groups were reduced iteratively until the sum of the daily emissionsequaled the mean emissions observed in E-41. The data from E-41 and theadjustment algorithm are given in Appendix 5.3-3.

5.3.6 Estimation of Basic Emission Rate for Near-zero Evap Vehicles

The basic emission rates for near-zero evap vehicles were estimated from the basicemission rates of enhanced evap vehicles. Similar to the methodology used toestimate BERs for enhanced evap vehicles, near-zero evap vehicles were assumedto emit like enhanced evap vehicles, but with emissions reduced by the ratio of thestandards. For PCs, the ratio is 0.5/2.0= 0.25. The BERs for vehicle classes otherthan passenger vehicles were determined using the ratio of the standards (relative toPCs) as outlined below. These ratios are applied to Normal and Moderate emittersonly.

Figure 5.3-2d. Estimated diurnal and resting loss emissions with respect to ambient tefor liquid leakers

Estimated Diurnal and Resting Emissions (Liquid Leakers)

0

2

4

6

8

10

12

14

16

0 20 40 60 80 100 120Temp (F)

g/hr

DiurnalResting

Class Specific Scaling Factor

Class Tech Near-zero Evap Standards Scalar (Ratio of Standards)PC Near-zero Evap 0.5 1T1 Near-zero Evap 0.65 1.3T2 Near-zero Evap 0.9 1.8T3 Near-zero Evap 1 2T4 Near-zero Evap 1 2T5 Near-zero Evap 1 2T6 Near-zero Evap 1 2T7 Near-zero Evap 1 2T8 Near-zero Evap 1 2

Phase-in Schedule

Near-zero evap vehicles are phased in as follows:

MY % Near-zero2004 402005 802006 100

5.3.7 Temperature and Fuel Correction Factor

The purpose of developing temperature and fuel correction factors is to adjustdiurnal and resting loss emission factors to other fuel and temperature conditions.Diurnal or resting loss emissions tend to increase when the ambient temperature ishigher. Likewise, diurnal or resting loss will rise if fuel RVP increases. To detectsuch a relationship, it is necessary to analyze a sample of vehicles undergoing thesame temperature cycle at different fuel RVP levels.

The data used in this portion of the analysis is a subset of the EPA database. Thissubset contains 8 CARB and 18 FI vehicles that were tested over the 72-96 Ftemperature cycle both at 7 and 9 RVP. Hourly HC averages were computed forthe temperature range between 72 to 96 F. As a result, we are able to discern therelationship of temperature and fuel RVP. Because of the small sample size, bothCARB and FI vehicles were combined for the data analysis. Finally, a regressionmodel was developed to depict the relationship as listed below.

f(T, RVP) = α(T+15) + β(RVP) + χ(T+15)(RVP) + Intercept (5.3-2)

where the domain of temperature (T) ranges from 55 to 110 F and RVP ranges from6.5 to 13 psi.

Therefore, the temperature and fuel correction factor

= f(T, RVP)/f(T, 9 RVP)

Table 5.3-3 lists the coefficients of the above equation and Figure 3 graphicallypresents the temperature and fuel correction equation. As expected, emissions riseas temperature or RVP increases. The temperature and fuel correction factor isapplicable to normals and moderates in FI and CARB categories.

5.3.8 Multi-day Correction Factors

Cumulative emissions of HC may be different for day 2 and day 3 of vehicle soak.Therefore, there is a need to develop a correction factor to account for the changesin total emissions for day 2 as well as day 3 and beyond.

Figure 5.3-3. Relationship between ambient temperature and fuel RVP on diurnaland resting evaporative loss

Table 5.3-3. Coefficient for Temperature and Fuel Correction Equation

TEMP FUEL TEMPFUEL INTERCEPTA B C D

-0.0822 -1.2507 0.0175 6.2152

f (T, RVP) = A*(T+15) + B*RVP + C*(T+15)*RVP + D (g/hr)

Temperature and Fuel Correction Factor = f(T, RVP)/f(T, 9)

where T and RVP ranges from 55 F to 110 F and 6.5 to 13 RVP respectively.

60 65 70 75 80 85 90 95 100 105 110 6

90

1

2

3

4

5

6

7

g/hr

Temp (F)

RVP (psi)

6-75-64-53-42-31-20-1

The data used for the multi-day correction factor analysis contains 101 vehiclestested at 9 psi over 72 hours. These vehicles were also tested over differenttemperature cycles; namely, 60 – 84 F, 65 – 105 F, 72 – 96 F, and 82 –106 F.

To develop multi-day correction factors, an analysis was performed to compare thecumulative HC emissions over 24 hours, 48 hours, and 72 hours, respectively. Itwas found that the temperature cycle has a minimal effect on the relative changes ofcumulative HC emissions for day 1, day 2 and day 3 for both CARB and FIvehicles. In other words, there were no relative changes in cumulative HC for day1, day 2, and day 3 when compared with various temperature cycles. Based uponcertification data on low emitting data, it is assumed that near-zero evap vehicleswill not have elevated emissions on the second and subsequent days.

On the other hand, the fuel-metering system appears to have a major impact on themulti-day correction factor. As shown in Figure 5.3-4, the hourly averageemissions for CARB vehicles remain almost constant throughout three dayswhereas hourly average emissions for FI vehicles increase daily. Table 5.3-4 liststhe multi-day correction factor for CARB and FI. Diurnal or resting loss for day 2as well as day 3 and beyond can be estimated by multiplying the multi-daycorrection factor by the basic emission rates of day 1.

5.3.9 Regime Growth Rates

The emission regime growth rates for CARB and FI were developed from thehistorical Air Resources Board condensed one-hour diurnal data, using 2g/test asthe cutpoint for normal and moderate categories. For consistency, it was assumed

F i g u r e 5 .3 - 4 . N o r m a l i z e d a v e r a g e t h r e e - d a y d i u r n a l a n d r e s t i n g e v a p o r a t i v e l o s s b a s e d o n 1 0 1 v e i c l e s a t 9 p s i R V P

T a b le 5 .3 - 4 . M u l t ip le d a y C o r r e c t io n F a c t o r s a t 9 p s i R V P

S y s te m D a y 1 D a y 2 D a y 3 +

C A R B 1 1 .0 1 1 .0 1F I 1 1 .5 3 1 .8 6E n h a n c e d 1 1 .0 0 1 .0 0N e a r Z e r o 1 1 .0 0 1 .0 0

A v e r a g e T h r e e - d a y D i u r n a l a n d R e s t i n g E m i s s i o n s

0 %

1 %

2 %

3 %

4 %

5 %

6 %

7 %

0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0H o u r

Perc

enta

ge o

f Tot

al E

mis

sion

s

C A R B ( n = 4 8 )F I ( n = 5 3 )

that the EPA’s assessment of the fraction of liquid leakers could be used for bothCARB and FI vehicles. Figure 5.3-5a and 5b present the regime growth rates of normal, moderate andliquid leakers for CARB and FI, respectively. The equations describing the regimegrowth pattern were also listed. As expected, CARB vehicles tend to attainmoderate and liquid leaker categories faster than FI vehicles. Moreover, thepercentage of liquid leakers for CARB is higher than FI, for any given age. In theevent that the sum of fractions exceed 100%, the regimes are normalized.

Figure 5.3-5a. Regime growth of normal, moderate, liquid leakers for CARB vehicles.

CARB EquationNormal Fraction = 0.92 + -0.0259*Age

Moderate Fraction = 0.085521 + 0.02468*(Age -1)LL Fraction = 0.06/(1+120*exp(-0.4*age))

Emission Regime Growth for CARB

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 5 9 13 17 21 25 29 33 37 41 45Age

Perc

ent

HighModerateNormal

As discussed in section 5.1, the growth of liquid leaking enhanced and near-zerofleets are assumed to be half of the FI growth rate.

5.3.10 I/M corrected Diurnal and Resting Loss Emission Factors

The California I/M program requires vehicles to undergo inspection biennially.Hence, we assume moderates will receive I/M benefit as some of the componentscausing high evaporative emissions are identified and repaired.

The average emission factor for normal emitters with respect to age is defined as follows:

Average EF for Normal Emitters in the Fleet (EF Ave Normal Emitters, Age) = Normal Emitter Growth Rate CARB *EF CARB*CARB Vehicle Fraction + Normal Emitter Growth Rate FI*EF FI*FI Vehicle Fraction (5.3-3)

Similarly, the average emission factor for moderate emitters with respect to age isdefined as follows:

Average EF for Moderate Emitters in the Fleet (EF Ave Moderate Emitters, Age)= Moderate Emitter Growth Rate CARB *EF CARB*CARB Vehicle Fraction +Moderate Emitter Growth Rate FI*EF FI*FI Vehicle Fraction

(5.3-4)

Figure 5.3-5b. Regime growth rate of normal, moderate, and liquid leakers for FI vehicles.

FI EquationNormal Fraction = 0.975 + -0.0189*Age

Moderate Fraction =0.0229 + 0.01821*(Age-1)LL Fraction = 0.06/(1+120*exp(-0.4*age))

Emission Regime Growth for FI

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1 5 9 13 17 21 25 29 33 37 41 45Age

Perc

ent

HighModerateNormal

Average EF for Liquid Leakers in the Fleet (EF Ave LL, Age)= Liquid Leaker Growth Rate CARB *EF CARB*CARB Vehicle Fraction +Liquid Leaker Growth Rate FI*EF FI*FI Vehicle Fraction (5.3-5)

Because of the I/M program, there is an emission benefit for moderate emitters. Inparticular, vehicles subject to I//M and successful repair will change their statusfrom moderate to normal emitters. Therefore, the moderate emitter growth rate forCARB and FI is adjusted accordingly.

Gas cap failure rates were well documented in a 1996 smog check study conductedby BAR. Therefore, gas cap failure rates were used to estimate the emission controlfailure rate in I/M. (See Appendix 5.3-4 for the methodology to estimate gas capfailure rate.)

It was assumed that the vehicles in the moderate emitter regime benefit from theI/M program. Specifically, vehicles in the moderate emitter category are identifiedand repaired, and will move to the normal emitter regime after repair. However,vehicles in the liquid leaker category were assumed not to be identified by I/M andthus will not change its emission regime size.

Fraction of moderate emitters moved to normal emitters per inspection period (RateModerate to Normal)

= Identification Rate (ID %)*Incremental Gas Cap Failure Rate (IGC Fail)*RepairEfficiency (Repair %) (5.3-6)

Thus, adjusted moderate emitter growth rate for both CARB and FI per inspectionperiod is as follows:= Moderate Emitter Growth Rate - Rate Moderate to Normal(5.3-7)

Assuming the identification rate and repair efficiency is 95%, the new moderateemitter growth rate is thus given as follows:

New Moderate Emitter Growth Rate = Moderate Emitter Growth Rate * (1- 0.95*gas cap failure rate) (5.3-8)

5.3.11 Moderate Emitter Growth Rate and OBDII

Because of the OBD II system, emissions control components are closely monitoredand likely to be repaired once malfunctioning components are detected. Therefore,OBDII vehicles will be modeled by suppressing the formation of moderate emittersfor the first seven years of a vehicle’s life. As a result, the new moderate emittergrowth rate for OBDII vehicles will be created by subtracting the fraction ofmoderate emitters for the first seven years, as presented below:

The adjusted moderate emitter regime growth

= 0.0229 + 0.01821*(Age-1) – 0.13216= 0.01821*(Age-1) – 0.10916 (5.3-9)

Note that if the fraction of moderate emitters becomes negative, it is consideredzero. It was assumed that the regime growth rate for liquid leakers would remainunchanged as listed in Figure 5.3-5b. Therefore, the fraction of normal emitters isgiven as follows:

Fraction of Normal Emitters = 1 – Adjusted Fraction of Moderate Emitters – Fraction of Liquid Leakers

5.3.12 Conclusions

While this study attempts to model both diurnal and resting loss using ambienttemperature as the driving force, there are limitations in modeling because of lackof data. Both diurnal and resting losses were modeled based on a definedtemperature cycle; however, these cycles may not faithfully depict the temperatureprofile experienced by vehicles under various initial conditions. Furthermore, withthe more stringent evaporative emission standards, newer vehicles are expected tohave less evaporative emission. Diurnal and resting loss emission data fromvehicles subject to the enhanced evaporative emission standards are needed.

Future studies should focus on the following issues:

(1) Vehicles of model year 1995 and beyond;(2) The effect of time and initial temperature on diurnal and resting loss;(3) Understanding the lag time between the peaks of emissions and ambient

temperature; (4) The impact of solar loading and wind on evaporative emissions; and(5) Comparing the diurnal and resting loss emissions with the methodology used by

the EPA.

The purpose of this analysis is to relate the one-hour condensed diurnal test to thediurnal and resting loss emissions over a 24-hour period. Thus, the corresponding24-hour cutpoint can be estimated from the 2 g/hr one-hour condensed diurnal test,similar to the cutpoint used in the hot soak analysis. The data used in thiscorrelation analysis came from the a study conducted by Automotive TestingLaboratories in 1994 (Contract No. A096-214). The conventional one-hour hotsoak results were compared with 24-hour test data. There were two diurnaltemperature profiles; namely, 60 to 84 F and 65 to 105 F. As expected, data fromthe temperature profile 65 to 105 F have a higher total emissions when compared tothe 60 to 84 F profile. Nevertheless, the correlation is developed in order toestimate a cutpoint to distinguish between normal and moderate emitters.

A p p e n d ix 5 .3 -1 . C o rre la t io n b e tw e e n o n e -h o u r c o n d e n s e d d iu rn a l a n d th e 2 4 -h o u r te s t ( 9 p s i R V P )

C o rre la t io n E q u a t io n :

2 4 -h o u r te s t(g ) = 1 1 .4 3 3 5 + 1 .9 1 9 5 * (O n e -h o u r c o n d e n s e d d iu rn a l)

0

1 0

2 0

3 0

4 0

5 0

6 0

7 0

8 0

9 0

0 2 4 6 8 1 0 1 2 1 4 1 6 1 8 2 0 2 2 2 4

O n e -h o u r c o n d e n s e d d iu r n a l (g )

24-h

our d

iurn

al a

nd re

stin

g lo

ss e

mis

sion

s (g

)

D a y 1 (6 0 -8 4 F )

D a y 1 (6 5 -1 0 5 F )

P re d ic te d

Appendix 5.3-2. Hourly average emissions based on emission status, model year groupings,fuel delivery system, and temperature cycles.

Status Normal Status NormalMY Group 1979 - 1994 MY Group Pre79

System CARB System CARBNumber 25 4 58 8 Number 4 1 7

Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F1 0.188 0.575 0.232 0.388 1 0.325 0.210 0.3642 0.218 0.557 0.245 0.437 2 0.211 0.238 0.3843 0.318 0.657 0.399 0.543 3 0.358 0.267 0.5324 0.428 0.729 0.598 0.855 4 0.613 0.386 0.7205 0.551 0.726 0.851 1.053 5 0.679 0.490 0.9316 0.750 0.810 1.123 1.275 6 0.785 0.639 1.2057 0.858 0.875 1.348 1.536 7 0.820 0.833 1.5308 0.837 1.024 1.286 1.406 8 0.787 0.869 1.6299 0.721 0.954 1.063 1.391 9 0.774 1.044 1.466

10 0.638 1.110 0.843 0.901 10 0.658 1.277 1.21811 0.513 0.999 0.615 0.886 11 0.505 1.273 0.95112 0.425 0.765 0.470 0.672 12 0.407 0.795 0.71613 0.380 0.575 0.410 0.602 13 0.334 0.755 0.65114 0.274 0.449 0.332 0.410 14 0.275 0.652 0.56315 0.209 0.290 0.275 0.307 15 0.330 0.425 0.53716 0.163 0.175 0.189 0.265 16 0.176 0.197 0.39117 0.151 0.107 0.184 0.252 17 0.217 0.183 0.41718 0.132 0.076 0.170 0.211 18 0.153 0.146 0.36019 0.127 0.095 0.149 0.187 19 0.166 0.135 0.36620 0.112 0.095 0.136 0.246 20 0.146 0.095 0.33221 0.115 0.039 0.118 0.150 21 0.102 0.088 0.29422 0.098 0.085 0.124 0.193 22 0.145 0.095 0.32623 0.104 0.037 0.116 0.190 23 0.141 0.026 0.30624 0.105 0.038 0.121 0.200 24 0.144 0.027 0.320

Sum 8.856 12.960 11.850 14.766 Sum 9.707 12.456 17.694

Status Normal Status NormalMY Group 1995 and beyond MY Group 1979 - 1994

System FI System FINumber 1 2 1 Number 118 17 173 77

Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F1 0.050 0.080 0.090 1 0.089 0.119 0.128 0.1462 0.040 0.079 0.112 2 0.113 0.119 0.149 0.1553 0.040 0.091 0.144 3 0.221 0.148 0.265 0.2404 0.060 0.109 0.203 4 0.361 0.196 0.391 0.3195 0.080 0.145 0.256 5 0.481 0.239 0.498 0.4416 0.090 0.166 0.314 6 0.540 0.302 0.603 0.5937 0.100 0.215 0.365 7 0.594 0.367 0.716 0.7968 0.100 0.220 0.347 8 0.480 0.429 0.689 0.9349 0.100 0.207 0.293 9 0.362 0.476 0.595 0.845

10 0.100 0.193 0.254 10 0.263 0.565 0.467 0.67711 0.110 0.155 0.156 11 0.176 0.674 0.338 0.46412 0.090 0.144 0.152 12 0.145 0.498 0.274 0.32813 0.080 0.126 0.111 13 0.118 0.426 0.222 0.27414 0.070 0.107 0.105 14 0.095 0.305 0.185 0.22215 0.060 0.091 0.081 15 0.080 0.194 0.139 0.18216 0.050 0.059 0.071 16 0.058 0.105 0.140 0.13617 0.050 0.073 0.067 17 0.054 0.090 0.108 0.11918 0.040 0.059 0.047 18 0.052 0.078 0.095 0.11519 0.040 0.054 0.057 19 0.042 0.075 0.087 0.10420 0.040 0.048 0.047 20 0.045 0.074 0.084 0.09221 0.030 0.049 0.049 21 0.041 0.058 0.072 0.09922 0.030 0.050 0.060 22 0.041 0.069 0.073 0.09023 0.030 0.043 0.046 23 0.043 0.054 0.075 0.08124 0.030 0.044 0.048 24 0.043 0.055 0.078 0.084

Sum 1.590 2.540 3.080 Sum 4.626 6.318 6.644 7.726

Appendix 5.3-2. (cont'd)

Status Moderate Status ModerateMY Group 1979 - 1994 MY Group Pre79

System CARB System CARBNumber 11 11 32 23 Number 11 15 31 7

Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F1 0.705 0.681 0.795 0.960 1 0.545 0.694 0.534 0.7742 0.825 0.961 0.924 1.193 2 0.740 0.883 2.292 0.8963 1.639 1.866 1.583 2.119 3 1.387 1.702 3.007 1.7934 2.509 2.839 2.285 3.277 4 1.802 2.068 3.865 3.4495 3.162 3.636 3.164 4.589 5 2.010 2.820 4.921 6.1076 3.689 4.517 4.031 6.239 6 2.351 3.474 6.371 9.1547 3.703 5.203 4.404 6.984 7 2.473 4.572 7.242 10.7388 3.046 5.350 4.005 6.355 8 2.426 5.251 6.797 9.4829 2.282 5.825 3.100 4.701 9 2.196 5.649 5.356 6.360

10 1.614 5.497 2.144 3.209 10 1.840 5.303 4.323 4.07911 1.133 4.279 1.469 2.062 11 1.507 4.503 2.894 2.43412 0.823 3.159 0.995 1.382 12 0.996 2.860 2.305 1.73613 0.703 2.196 0.869 1.014 13 0.784 1.651 2.135 1.25614 0.630 1.582 0.680 0.789 14 0.673 0.980 1.743 0.85015 0.480 1.240 0.551 0.647 15 0.565 0.659 1.512 0.64616 0.357 0.652 0.429 0.433 16 0.433 0.498 1.176 0.47117 0.353 0.661 0.381 0.439 17 0.398 0.359 1.286 0.40518 0.341 0.667 0.339 0.392 18 0.393 0.364 1.213 0.44219 0.276 0.602 0.306 0.333 19 0.388 0.320 1.105 0.30020 0.263 0.436 0.290 0.318 20 0.341 0.263 1.041 0.30321 0.252 0.286 0.274 0.245 21 0.335 0.225 0.942 0.30922 0.237 0.339 0.226 0.253 22 0.277 0.206 0.979 0.23823 0.211 0.295 0.246 0.233 23 0.320 0.202 0.930 0.22924 0.214 0.302 0.253 0.240 24 0.324 0.204 0.973 0.237

Sum 30.316 56.572 34.036 48.015 Sum 26.534 48.463 68.335 62.278

Status Moderate Status HighMY Group 1979 - 1994 MY Group All

System FI System AllNumber 13 2 36 55 Number 2 35 2

Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F Hour 60-84 F 65 - 105 F 72 -96 F 82 -106 F1 0.682 0.320 0.620 0.378 1 2.895 0.814571 3.752 0.829 0.318 0.901 0.711 2 3.095 3.887198 4.354863 1.660 0.380 1.821 1.708 3 4.015 4.491217 6.2603414 2.539 0.504 3.029 2.748 4 4.915 5.457016 8.395 3.192 0.706 4.143 3.711 5 5.695 6.350595 10.29566 3.679 0.918 5.069 4.820 6 6.275 7.772331 13.653827 3.440 1.441 5.403 5.700 7 6.35 8.755185 11.597318 2.294 1.905 4.879 5.053 8 6.185 8.919793 12.677549 1.731 2.233 3.454 3.702 9 5.5 7.698219 9.868606

10 1.160 2.525 2.197 2.450 10 5.075 6.860281 8.37266311 0.900 2.207 1.349 1.381 11 4.805 5.127289 7.31599212 0.886 2.133 0.894 0.876 12 4.52 4.342946 5.65022213 0.677 1.800 0.602 0.637 13 3.635 4.175853 5.53425914 0.574 1.519 0.336 0.488 14 3.13 3.764295 4.39658515 0.552 1.085 0.239 0.388 15 2.86 3.547298 3.8859316 0.451 0.729 0.179 0.279 16 2.785 2.836253 3.93503617 0.418 0.650 0.200 0.273 17 2.735 3.121084 3.73644118 0.389 0.576 0.206 0.224 18 2.625 3.101931 3.57812919 0.313 0.447 0.154 0.194 19 2.53 2.844746 3.63748620 0.258 0.486 0.156 0.173 20 2.54 2.763386 3.62070121 0.252 0.332 0.152 0.161 21 2.675 2.563845 3.67936922 0.258 0.349 0.173 0.169 22 2.535 2.659463 3.7126423 0.270 0.282 0.166 0.162 23 2.505 2.494234 3.45785724 0.273 0.292 0.171 0.165 24 2.505 2.607179 3.549976

Sum 28.830 27.015 36.987 36.630 Sum 92.385 106.956 148.911

Appendix 5.3-3 - Enhanced Evap VehicleAl ith

CRC Project E-41World Evaporative Testing of Late Model In-Use Vehicles

24-hourDHB

Veh # Yr./Make/Model Gramsenh E-41008 1997 Dodge Stratus 0.38

2.4L, PFI, 16.0 tank

enh E-41020 1997 Plymouth Breeze 0.592.0L, PFI, 16.0 tank

enh E-41027 1996 Dodge Caravan 0.822.4L, PFI, 20.0 tank

enh E-41032 1996 Chevrolet S-10 0.354.3L, PFI, 19.0 tank

mean enh 0.54

Spreadsheet to calculate enhanced evap vehicle reductionHour Temp ER 9RVP RVPCF ER 7RVP E41

profile N/79-94 9-->7 RVP N/79-94 Enhanced1 65.0 0.1548 0.6962 0.1077 0.01232 68.6 0.2075 0.6611 0.1372 0.01573 72.3 0.2557 0.6386 0.1633 0.01874 75.9 0.3011 0.6229 0.1876 0.02155 79.5 0.3457 0.6113 0.2114 0.02426 83.2 0.3914 0.6025 0.2358 0.02707 86.8 0.4401 0.5955 0.2621 0.03008 90.5 0.4937 0.5898 0.2912 0.03339 94.1 0.5541 0.5851 0.3242 0.0371

10 97.7 0.6232 0.5811 0.3621 0.041411 101.4 0.7028 0.5777 0.4060 0.046512 105.0 0.7950 0.5748 0.4570 0.052313 101.7 0.5454 0.5775 0.3150 0.036014 98.3 0.4599 0.5805 0.2670 0.030615 95.0 0.3824 0.5840 0.2233 0.025616 91.7 0.3128 0.5881 0.1840 0.021117 88.3 0.2512 0.5930 0.1489 0.017018 85.0 0.1975 0.5988 0.1182 0.013519 81.7 0.1518 0.6059 0.0920 0.010520 78.3 0.1140 0.6148 0.0701 0.008021 75.0 0.0842 0.6264 0.0527 0.006022 71.7 0.0623 0.6418 0.0400 0.004623 68.3 0.0484 0.6635 0.0321 0.003724 65.0 0.0425 0.6962 0.0296 0.0034 factor =

enh/(79-94)7.9176 4.7185 0.5400 0.1144

Appendix 5.3-4 Estimation of the gas cap failure rate from BAR's smog check data performed in Spring 1996.

MY Age Observed Failure Rate Predicted Failure70 26.25 25.0% 34.2%71 25.25 40.0% 34.0%72 24.25 21.4% 33.7%73 23.25 36.1% 33.3%74 22.25 35.0% 32.7%75 21.25 43.3% 31.9%76 20.25 30.6% 30.7%77 19.25 35.1% 29.2%78 18.25 25.2% 27.3%79 17.25 27.6% 24.8%80 16.25 17.1% 22.0%81 15.25 16.7% 18.8%82 14.25 13.2% 15.6%83 13.25 9.7% 12.4%84 12.25 9.8% 9.6%85 11.25 9.7% 7.2%86 10.25 7.3% 5.3%87 9.25 5.6% 3.8%88 8.25 5.2% 2.7%89 7.25 4.3% 1.9%90 6.25 3.2% 1.3%91 5.25 1.4% 0.9%92 4.25 1.6% 0.6%93 3.25 1.0% 0.4% Parameter Value Model94 2.25 1.3% 0.3% K 0.347 Y= K/(1+((K-No)/No)*EXP(-R*AGE))95 1.25 na 0.2% No 0.0013396 0.25 na 0.1% R 0.3758

Predicted Gas Cap Failure Rates vs Raw Data

-5%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

0 5 10 15 20 25 30

Age

Perc

enta

ge

Observed Failure RatePredicted Failure